Amiodarone is a potent antiarrhythmic agent and can cause potentially life-threatening pulmonary fibrosis. Of the numerous side effects associated with amiodarone therapy, lugn toxicity is one of the most serious adverse reactions. Recently, we experienced a case of amiodarone-induced pulmonary toxicity (APT), which induced severe dyspnea and productive coughing, confirmed by cytologic and electron microscopic examination of the bronchoalveolar lavage (BAL). The symptoms and abnormalities in the chest X-ray were improved after the withdrawal of amiodarone. Cytologic examination of the BAL revealed numerous foam cells with cytoplasmic vacuoles or small particles. Ultrastructurally, the foam cells demonstrated characteristic lysosomal inclusions, which were electron-dense multilamellated bodies, crystalloid bodies, and mixed forms with small lipid vacuoles. It is strongly suggested that only cytologic and electron microscopic examination of the BAL without open lung biopsy is enough for diagnosis of APT, when APT is clinically suspected in a patient who has a history or ingestation of amiodarone.
Amiodarone ; Biopsy ; Bronchoalveolar Lavage* ; Cough ; Cytoplasm ; Diagnosis ; Dyspnea ; Foam Cells ; Humans ; Lung ; Microscopy, Electron ; Pulmonary Fibrosis ; Thorax ; Vacuoles

Objectives: ZZThe aim was to find effective vectorization and classification models to predict a psychiatric diagnosis from text-basedmedical records. Methods: ZZElectronic medical records (n = 494) of present illness were collected retrospectively in inpatient admission notes withthree diagnoses of major depressive disorder, type 1 bipolar disorder, and schizophrenia. Data were split into 400 training data and 94 independentvalidation data. Data were vectorized by two different models such as term frequency-inverse document frequency (TF-IDF)and Doc2vec. Machine learning models for classification including stochastic gradient descent, logistic regression, support vectorclassification, and deep learning (DL) were applied to predict three psychiatric diagnoses. Five-fold cross-validation was used to find aneffective model. Metrics such as accuracy, precision, recall, and F1-score were measured for comparison between the models. Results: ZZFive-fold cross-validation in training data showed DL model with Doc2vec was the most effective model to predict the diagnosis(accuracy = 0.87, F1-score = 0.87). However, these metrics have been reduced in independent test data set with final workingDL models (accuracy = 0.79, F1-score = 0.79), while the model of logistic regression and support vector machine with Doc2vec showedslightly better performance (accuracy = 0.80, F1-score = 0.80) than the DL models with Doc2vec and others with TF-IDF. Conclusions ZZThe current results suggest that the vectorization may have more impact on the performance of classification thanthe machine learning model. However, data set had a number of limitations including small sample size, imbalance among the category,and its generalizability. With this regard, the need for research with multi-sites and large samples is suggested to improve the machinelearning models.

Objectives: ZZThe aim was to find effective vectorization and classification models to predict a psychiatric diagnosis from text-basedmedical records. Methods: ZZElectronic medical records (n = 494) of present illness were collected retrospectively in inpatient admission notes withthree diagnoses of major depressive disorder, type 1 bipolar disorder, and schizophrenia. Data were split into 400 training data and 94 independentvalidation data. Data were vectorized by two different models such as term frequency-inverse document frequency (TF-IDF)and Doc2vec. Machine learning models for classification including stochastic gradient descent, logistic regression, support vectorclassification, and deep learning (DL) were applied to predict three psychiatric diagnoses. Five-fold cross-validation was used to find aneffective model. Metrics such as accuracy, precision, recall, and F1-score were measured for comparison between the models. Results: ZZFive-fold cross-validation in training data showed DL model with Doc2vec was the most effective model to predict the diagnosis(accuracy = 0.87, F1-score = 0.87). However, these metrics have been reduced in independent test data set with final workingDL models (accuracy = 0.79, F1-score = 0.79), while the model of logistic regression and support vector machine with Doc2vec showedslightly better performance (accuracy = 0.80, F1-score = 0.80) than the DL models with Doc2vec and others with TF-IDF. Conclusions ZZThe current results suggest that the vectorization may have more impact on the performance of classification thanthe machine learning model. However, data set had a number of limitations including small sample size, imbalance among the category,and its generalizability. With this regard, the need for research with multi-sites and large samples is suggested to improve the machinelearning models.

No abstract available.

BACKGROUND: Insular thyroid carcinoma (ITC) is a relatively infrequent thyroid carcinoma that has distinctive histologic features. ITC shows an aggressive clinical course and the predominant presence of an insular component, which has been reported to be an independent factor of a poor prognosis. We retrospectively examined clinical details of the nine ITC patients, which represented 9 years of experience with ITC, and investigated the expressions of variable neuroendocrine and other immunohistochemical markers associated with well-differentiated thyroid carcinomas. METHODS: We adopted an immunohistochemical approach and studied the expressions of synaptophysin, chromogranin A, CD56, NSE, S-100, RET, PPARgamma, calcitonin, galectin-3, and thyroglobulin in formalin-fixed, paraffin embedded tissue array slides of the 9 ITC patients, and investigated clinical features. Seven cases of follicular carcinoma and 4 cases of medullary carcinoma were also included as controls. RESULTS: ITCs were positive for synaptophysin (44%, 4/9), CD56 (11%, 1/9), NSE (89%, 8/9), S100 (67%, 6/9), calcitonin (22%, 2/9), galectin-3 (78%, 7/9), and thyroglobulin (100%, 9/9), but completely negative for chromogranin A, RET, and PPARgamma. CONCLUSION: ITCs express neuroendocrine markers in variable proportions and appear not to be associated with the oncoproteins of conventional thyroid carcinomas. Notably, its differential diagnosis from medullary carcinoma is required in cases showing focal calcitonin positivity.
Calcitonin ; Carcinoma, Medullary ; Chromogranin A ; Diagnosis, Differential ; Galectin 3 ; Humans ; Oncogene Proteins ; Paraffin ; PPAR gamma ; Prognosis ; Retrospective Studies ; Synaptophysin ; Thyroglobulin ; Thyroid Gland* ; Thyroid Neoplasms

PURPOSE: Inter-hospital transport poses a number of challenging issues, including prolonging the time interval from symptoms to optimal reperfusion therapy after ischemic stroke. It is unclear whether urbanization is associated with outcomes of inter-hospital transfer including length of stay at the referring hospital (D1LOS). METHODS: A prospective stroke registry from 23 Emergency Departments (ED) from 2007 to 2012 over the nation was collected. Ischemic stroke patients who arrived at the first ED within 24 hours of onset (S2D1) were enrolled. Patients were excluded if time intervals or address were incorrect or missing. Main exposure was urbanization level; urban > or =10,000 and rural <10,000 population. Primary outcome was D1LOS. The secondary outcomes were symptoms to door of the first ED (S2D1) and transfer time to the final ED (T2D2). We compared the D1LOS, S2D1, and T2D2 with median and inter-quartile range (IQR) by urbanization level. RESULTS: Of 5,909 patients transferred from other hospitals, 2,289 patients were analyzed; 1,441 (63%) patients in urban areas, 848 (37%) patients in rural areas were included. The D1LOS and S2D1 in urban was longer than those in rural; 100 minutes (IQR 50~208) for urban VS 82.5 minutes (IQR 48~170.5) for rural (p=0.01) and 66 minutes (IQR 30~240) for urban VS 90 minutes (IQR 30~330) for rural (p=0.001). T2D2 in urban was shorter than that in rural; 54 minutes (IQR 36~78), 40 minutes (IQR 25~65) (p< or =0.00), respectively. CONCLUSION: Urban EDs showed longer D1LOS before transferring patients to the hospital for definite care. Strategy for reducing delay due to inter-hospital transport should differ according to urbanization.
Emergency Medical Services ; Emergency Service, Hospital ; Humans ; Length of Stay ; Patient Transfer ; Reperfusion ; Stroke* ; Urbanization*

Recently, the numbers of lung transplantation (LT) has been increased in Korea. However, post-LT outcome has not been successful in all patients, which may be partially affected by the primary lung disease. Therefore comprehensive understanding in original pathological diagnosis of patients with LT would be needed for achieving better clinical outcome. To address this issue, we performed clinico-pathological analysis of the explanted lungs from 29 patients who underwent LT over a 9-yr period in Seoul National University Hospital. Among them, 26 patients received single (1/26) or double (25/26) LT, while heart-lung transplantation was performed in 3 patients. The final clinico-pathological diagnoses were idiopathic pulmonary fibrosis/usual interstitial pneumonia (UIP) (n = 6), acute interstitial pneumonia (AIP)/diffuse alveolar damage (DAD) (n = 4), AIP/non-specific interstitial pneumonia with DAD (n = 1), collagen vascular disease-related interstitial lung disease (CVD-ILD)/DAD (n = 3), CVD-ILD/UIP (n = 1), lymphangioleiomyomatosis (n = 1), bronchiectasis (n = 4), pulmonary arterial hypertension (n = 2), tuberculosis (n = 1), bronchiolitis obliterans (BO) (n = 1), and lung cancer (n = 1). Moreover, 4 patients who had chemotherapy and hematopoietic stem cell transplantation due to hematologic malignancy showed unclassifiable interstitial pneumonia with extensive fibrosis in the lungs. Our study demonstrates that pathology of the explanted lungs from Korean patients with LT is different from that of other countries except for interstitial lung disease and bronchiectasis, which may be helpful for optimization of selecting LT candidates for Korean patients.
Acinetobacter baumannii/isolation & purification ; Adolescent ; Adult ; Aged ; Bronchiectasis/*pathology ; Child ; Female ; Humans ; Idiopathic Pulmonary Fibrosis/*pathology ; Lung/microbiology/*pathology ; Lung Diseases, Interstitial/*pathology ; *Lung Transplantation ; Male ; Middle Aged ; Republic of Korea ; Treatment Outcome ; Vancomycin-Resistant Enterococci/isolation & purification ; Young Adult

Pellino-1 is a ubiquitin (Ub) E3 ligase that plays a role in M1, but not M2a polarization of macrophages. However, it is unknown whether Pellino-1 regulates IL-10-mediated M2c polarization of macrophages. Here, we found that Pellino-1 attenuated tumor growth by inhibiting M2c polarization of macrophages. Upon IL-10 stimulation, Pellino-1-deificient bone marrow-derived macrophages (BMDMs) showed higher expression of M2c markers, but not M2a, and M2b markers than wild-type (WT) BMDMs, indicating that Pellino-1 inhibits M2c polarization of macrophages. Pellino-1-deficient BMDMs exhibited a defect in mitochondria respiration, but enhancement of glycolysis during M2c polarization. During M2c polarization of macrophages, Pellino-1 increased STAT1 phosphorylation via K63-linked ubiquitination of IL-1 receptor associated kinase 1 (IRAK1). Furthermore, Lysm-CrePellino-1(fl/fl) mice showed enhancement of tumor growth via regulating M2c polarization of tumor-associated macrophages. These results demonstrate that Pellino-1 inhibits IL-10-induced M2c macrophage polarization via K63-linked ubiquitination of IRAK1 and activation of STAT1, thereby inhibiting tumor growth in vivo.
Animals ; Glycolysis ; Interleukin-1 ; Interleukin-10 ; Macrophages ; Mice ; Mitochondria ; Phosphorylation ; Phosphotransferases ; Respiration ; Ubiquitin ; Ubiquitin-Protein Ligases ; Ubiquitination

BACKGROUND: Xanthine derivatives have been used to treat a variety of medical conditions including respiratory disease and neural degeneration. However, few studies have reported their effects on bone regeneration. Therefore, we investigated the effects of KPR-A148, a synthetic xanthine derivative on osteoblast differentiation in vitro and bone regeneration in vivo. METHODS: The cytotoxicity of KPR-A148 was evaluated using MC3T3-E1 cells by the 3-(4,5-dimethylthiazolyl-2)-2,5-diphenyltertrazolium bromide assay. The effects of KPR-A148 on osteoblast differentiation were examined by alkaline phosphatase staining, Alizarin red S staining, and real-time PCR of osteoblast differentiation marker genes. To investigate the effects of KPR-A148 on in vivo bone regeneration, a KPR-A148-containing collagen sponge was implanted into a mouse calvarial defect and KPR-A148 was injected twice, weekly. Bone regeneration was evaluated quantitatively by micro-CT and qualitatively by hematoxylin and eosin, as well as Masson's Trichrome staining. RESULTS: KPR-A148 did not show toxicity in the MC3T3-E1 cells and promoted osteoblast differentiation in a concentration-dependent manner. 10 µM of KPR-A148 showed the most significant effect on alkaline phospatase staining and matrix mineralization. KPR-A148 increased the expression of osteoblast marker genes in both the early and late stages of differentiation. In addition, KPR-A148 significantly induced new bone formation in the calvarial defect model. CONCLUSION: These results demonstrate that KPR-A148 strongly induces osteoblast differentiation and new bone formation. Therefore, it could be used as a potential therapeutic agent for regenerating bone following its destruction by disease or trauma.
Alkaline Phosphatase ; Animals ; Bone Regeneration ; Collagen ; Eosine Yellowish-(YS) ; Hematoxylin ; In Vitro Techniques ; Mice ; Miners ; Osteoblasts ; Osteogenesis ; Porifera ; Real-Time Polymerase Chain Reaction ; Xanthine